Geographic information data dynamic encryption method and system based on spatial-temporal characteristics

Abstract

The invention provides a geographic information data dynamic encryption method and system based on spatial-temporal characteristics. The method comprises the following steps: dynamically identifying a hot spot region by analyzing a timestamp, geographic coordinates and frequency data of a user access request, and generating a multi-dimensional access weight matrix fusing a time attenuation factor and a spatial distance factor; and dividing data block granularity according to the access popularity value in the matrix, and performing spatial segmentation on the geographic information data. A dynamic key seed is generated in combination with a space-time factor, a temporary encryption key is obtained through discrete logarithm operation, homomorphic encryption is implemented, and a space-time feature identifier is embedded. And during cross-mechanism aggregation, the authority is verified through the identifier, ciphertext superposition is completed, and secure data aggregation is realized. According to the method, through spatio-temporal dynamic encryption and authority verification, secure sharing and efficient aggregation of multi-mechanism geographic information data are realized.

Description

Technical Field

[0001] This application relates to the field of data encryption technology, and in particular, to a dynamic encryption method and system for geographic information data based on spatio-temporal features. Background Art

[0002] In the scenario of multi-institutional geographic information data sharing, the access requests of different users (such as government departments, enterprises, research institutions, etc.) have significant spatio-temporal dynamics, manifested as strong correlations among access frequency, geographical location, and time. To ensure data security, dynamic permission control needs to be implemented, that is, adjusting the data encryption policy according to the real-time access behavior of users (such as hot spots, time distribution), and at the same time supporting cross-institutional ciphertext aggregation calculation. This scenario requires the encryption method to have spatio-temporal perception ability, dynamic key generation mechanism, and fine-grained access control to balance data security and sharing efficiency.

[0003] Currently, a representative solution for this requirement is the method based on fixed grid partitioning combined with attribute-based encryption. This method divides geographic information data into fixed grids, combines user attributes (such as institutions, roles) to generate access policies, and uses attribute-based encryption method to achieve data encryption and permission control. When sharing data, the system verifies user permissions through predefined attribute policies, and only allows eligible users to decrypt specific grid data.

[0004] Although this solution can achieve basic permission control, it has the following problems: Mismatch between static partitioning and dynamic access requirements: Fixed grids cannot adapt to the spatio-temporal changes of access hot spots, resulting in inefficient encryption or over-authorization. Rigid permission policy: The attribute-based encryption method relies on predefined attributes and cannot dynamically adjust keys and permissions according to real-time access behavior (such as sudden high-frequency requests). Lack of spatio-temporal perception security mechanism: It does not consider time decay and spatial distance factors, making it difficult to prevent attacks based on historical keys or location impersonation, and permission conflicts are likely to occur during cross-institutional data aggregation.

[0005] In summary, the existing solutions are difficult to meet the dynamic security requirements in the multi-institutional geographic information sharing scenario, and there is an urgent need for a dynamic encryption and permission control method that integrates spatio-temporal features. Summary of the Invention

[0006] This application provides a dynamic encryption method and system for geographic information data based on spatio-temporal features to solve the problems of poor encryption effect and low security of geographic information data in the prior art.

[0007] In the first aspect, this application provides a dynamic encryption method for geographic information data based on spatio-temporal features, including: Determining a dynamic hot spot area according to the obtained timestamp data, geographic coordinate data, and access frequency data of user access requests in a multi-institutional sharing scenario; Generate a multi-dimensional access weight matrix based on the time distribution density and spatial coverage radius of the dynamic hot spot area, where the multi-dimensional access weight matrix includes a time dimension decay factor and a spatial dimension distance factor; Analyze the access heat values of the spatial grids in the multi-dimensional access weight matrix, determine the corresponding block granularity levels according to the numerical ranges of the access heat values, and perform spatial segmentation operations on the target geographic information data according to the block granularity levels to generate multiple data blocks matching the spatial grids; Generate a dynamic key seed based on the time dimension decay factor and the spatial dimension distance factor; Perform discrete logarithm operations on the dynamic key seed to generate a temporary encryption key, use the temporary encryption key to perform homomorphic encryption operations on the multiple data blocks, and embed spatio-temporal feature identifiers associated with the multi-dimensional access weight matrix in the encrypted data blocks; When receiving a cross-institutional data aggregation request, extract the permission level information in the spatio-temporal feature identifier from the encrypted data blocks, perform ciphertext superposition operations on the multiple encrypted data blocks based on the homomorphic encryption characteristics, and simultaneously verify the matching degree between the user identity identifier of the requester and the permission level information to complete the secure aggregation operation.

[0008] Optionally, the analyzing the access heat values of the spatial grids in the multi-dimensional access weight matrix, determining the corresponding block granularity levels according to the numerical ranges of the access heat values, and performing spatial segmentation operations on the target geographic information data according to the block granularity levels to generate multiple data blocks matching the spatial grids includes: Mark the initial access heat value within the coverage range of each spatial grid in the multi-dimensional access weight matrix, where the initial access heat value is the product of the time distribution density and the spatial coverage radius; Construct a heat diffusion window based on the adjacent relationship between spatial grids, and perform horizontal propagation and accumulation of the initial access heat value within the heat diffusion window to obtain the access heat value; Input the access heat value into a preset discretized step function to divide the corresponding block granularity levels according to each step interval through the discretized step function; Determine the maximum number of blocks allowed within the spatial grid according to the block granularity level, and perform radial cutting at equal angular intervals along the longitude and latitude directions with the center point of the spatial grid as the reference to form multiple initial data blocks; When the multiple initial data blocks generate overlapping areas with the cutting lines of adjacent spatial grids, re-divide the boundaries of each initial data block with the center line of the overlapping area as the boundary to generate multiple data blocks, and each data block belongs to a single spatial grid.

[0009] Optionally, generating a dynamic key seed based on the time dimension attenuation factor and the space dimension distance factor includes: Performing a time stamp difference operation based on the time dimension attenuation factor to generate a time attenuation coefficient; Performing a geographic coordinate difference operation based on the space dimension distance factor to generate a space distance coefficient; Inputting the time attenuation coefficient and the space distance coefficient into a non-linear transformation function to generate a dynamic key seed.

[0010] Optionally, performing a discrete logarithm operation on the dynamic key seed to generate a temporary encryption key, and performing a homomorphic encryption operation on the multiple data blocks using the temporary encryption key, and embedding a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix in the encrypted data block, including: Performing a circular shift operation on the initial value of the dynamic key seed, performing an exclusive OR operation on each element corresponding to the initial value and the adjacent element after shifting to generate an extended key sequence, and selecting elements that meet a preset parity condition according to the parity distribution characteristics of the elements in the extended key sequence to generate a temporary encryption key; Dividing the temporary encryption key into multiple sub-key segments, each sub-key segment corresponding to the encryption position of a data block, calculating the starting offset of the sub-key segment in the temporary encryption key sequence according to the spatial grid coordinates of the data block, and intercepting a sub-key segment of the corresponding length according to the offset as the encryption parameter of the data block to perform a homomorphic encryption operation on the multiple data blocks to generate encrypted data blocks; Performing bit cross-weaving on the time distribution density feature code and the space coverage radius feature code of the spatial grid from the multi-dimensional access weight matrix to generate a spatio-temporal feature identifier, and embedding the spatio-temporal feature identifier into the corresponding encrypted data block.

[0011] Optionally, when receiving a cross-institutional data aggregation request, extracting the permission level information in the spatio-temporal feature identifier, performing a ciphertext superposition operation on multiple encrypted data blocks based on the homomorphic encryption property, and verifying the matching degree between the user identity identifier of the requestor and the permission level information to complete the secure aggregation operation, including: When receiving a cross-institutional data aggregation request, extracting the spatio-temporal feature identifier from the encrypted data block, and extracting the permission level information composed of a digital sequence from a fixed position of the spatio-temporal feature identifier, comparing the permission level information item by item with a pre-stored institutional permission mapping table, and screening out multiple target encrypted data blocks that are allowed to participate in the superposition operation from the multiple encrypted data blocks; The user identity identifier of the requester carried in the cross-institutional data aggregation request is split into a header verification code and a tail permission identifier. The header verification code is converted into a binary mask through a preset permutation rule and performs bitwise overlay matching with the verification area of the spatio-temporal feature identifier. The tail permission identifier reversely derives a theoretical identity feature identifier through the permission level information. When the number of consecutive matches between the theoretical identity feature identifier and the verification bit of the user identity identifier reaches a preset threshold, the overlay operation channel is activated. The multiple target encrypted data blocks are separately split into multiple ciphertext segments, and the ciphertext segments with the same position index in the multiple encrypted data blocks are input into the overlay operation channel. A dynamic weight factor is generated based on the permission level information corresponding to the multiple encrypted data blocks, and the ciphertext segments within the same channel are subjected to weighted multiplication and accumulation to generate the ciphertext segment corresponding to the position in the aggregated ciphertext. During the overlay process, the number of ciphertext segments already overlaid in the aggregated ciphertext for each encrypted data block is statistically calculated in real time. If the overlay quantity of any target encrypted data block exceeds the maximum overlay segment number threshold corresponding to the permission level information of the target encrypted data block, the overlay operation channel corresponding to the target encrypted data block is closed.

[0012] Optionally, performing a time stamp difference operation based on the time dimension attenuation factor to generate a time attenuation coefficient, including: Recording the first marked time stamp value at the starting moment of the acquisition event and the second marked time stamp value at the current moment of the event using the same time measurement rule; Performing a time stamp difference operation on the first marked time stamp value and the second marked time stamp value to obtain an initial time difference, and dividing the initial time difference quantity by the corresponding value of the preset basic time unit to obtain a standard time interval quantity; Extracting a predefined time dimension attenuation factor, using the time dimension attenuation factor as the base number, using the standard time interval quantity as the exponent, and performing the exponentiation operation of the base number to generate a time attenuation coefficient.

[0013] Optionally, bitwise cross-weaving the time distribution density feature code and the space coverage radius feature code of the space grid from the multi-dimensional access weight matrix to generate a spatio-temporal feature identifier, and embedding the spatio-temporal feature identifier into the corresponding encrypted data block, including: Extracting the time distribution density feature code from the first predefined column of the multi-dimensional access weight matrix and extracting the space coverage radius feature code from the second predefined row; Converting the time distribution density feature code and the space coverage radius feature code into binary sequences of a fixed length respectively, and sequentially cross-joining each bit of the two binary sequences according to the alternating arrangement rule to generate a spatio-temporal feature identifier; Reserve an embedded area of a fixed length at the head of the encrypted data block, and fill the spatio-temporal feature identifier into the embedded area bit by bit, covering the values of the original data bits.

[0014] In a second aspect, the present application provides a dynamic encryption system for geographic information data based on spatio-temporal features, including: A determination module, configured to determine a dynamic hot spot area according to the timestamp data, geographic coordinate data, and access frequency data of a user access request in a multi-institution sharing scenario; A first generation module, configured to generate a multi-dimensional access weight matrix according to the time distribution density and spatial coverage radius of the dynamic hot spot area, where the multi-dimensional access weight matrix includes a time dimension decay factor and a spatial dimension distance factor; A second generation module, configured to parse the access heat value of the spatial grid in the multi-dimensional access weight matrix, determine the corresponding block granularity level according to the numerical range of the access heat value, and perform a spatial segmentation operation on the target geographic information data according to the block granularity level to generate multiple data blocks matching the spatial grid; A third generation module, configured to generate a dynamic key seed based on the time dimension decay factor and the spatial dimension distance factor; An encryption module, configured to perform a discrete logarithm operation on the dynamic key seed to generate a temporary encryption key, use the temporary encryption key to perform a homomorphic encryption operation on the multiple data blocks, and embed a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix in the encrypted data block; A processing module, configured to, when receiving a cross-institution data aggregation request, extract the permission level information in the spatio-temporal feature identifier from the encrypted data block, perform a ciphertext superposition operation on the multiple encrypted data blocks based on the homomorphic encryption feature, and verify the matching degree between the user identity identifier of the requestor and the permission level information to complete a secure aggregation operation.

[0015] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a dynamic encryption method for geographic information data based on spatio-temporal features as described in the first aspect above.

[0016] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, where when the computer program is executed by a computer, it implements a dynamic encryption method for geographic information data based on spatio-temporal features as described in the first aspect.

[0017] In the embodiments of the present application, based on the obtained timestamp data, geographical coordinate data, and access frequency data of user access requests in a multi-institutional sharing scenario, a dynamic hot spot area is determined. By analyzing the timestamp, geographical coordinates, and access frequency, the spatio-temporal area with a high access frequency is identified in real time, ensuring that the encryption policy dynamically matches the user behavior and improving the accuracy of security control.

[0018] A multi-dimensional access weight matrix is generated according to the time distribution density and spatial coverage radius of the dynamic hot spot area. The multi-dimensional access weight matrix includes a time dimension decay factor and a spatial dimension distance factor. By combining the time decay factor and the spatial distance factor, the spatio-temporal characteristics of access behavior are quantified, providing a computable weight basis for dynamic encryption and enhancing the adaptability of key generation.

[0019] The access heat value of the spatial grid in the multi-dimensional access weight matrix is analyzed. According to the numerical range of the access heat value, the corresponding block granularity level is determined. The target geographical information data is subjected to a spatial segmentation operation according to the block granularity level, generating multiple data blocks that match the spatial grid. The data block granularity is adaptively adjusted according to the access heat. The hot spot area is segmented with a finer granularity, optimizing the storage and calculation efficiency and reducing the redundant overhead in the non-hot spot area.

[0020] Based on the time dimension decay factor and the spatial dimension distance factor, a dynamic key seed is generated. The dynamic key seed is generated based on spatio-temporal factors, ensuring that the key is updated as the access behavior changes and preventing security risks caused by key reuse or static leakage.

[0021] A discrete logarithm operation is performed on the dynamic key seed to generate a temporary encryption key. The temporary encryption key is used to perform a homomorphic encryption operation on the multiple data blocks, and a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix is embedded in the encrypted data blocks. The irreversibility of the key is enhanced through the discrete logarithm operation, and homomorphic encryption is combined to support ciphertext calculation, meeting the cross-institutional data aggregation requirements. At the same time, the spatio-temporal identifier is embedded to achieve fine-grained permission control.

[0022] When a cross-institutional data aggregation request is received, the permission level information in the spatio-temporal feature identifier is extracted from the encrypted data blocks. Based on the characteristics of homomorphic encryption, a ciphertext superposition operation is performed on the multiple encrypted data blocks. At the same time, the matching degree between the user identity identifier of the requestor and the permission level information is verified to complete the secure aggregation operation. The requestor's permission is verified through the spatio-temporal feature identifier, and the encrypted data is directly superimposed using homomorphic encryption, ensuring the security and efficiency of the aggregation process and avoiding the risk of plaintext exposure.

[0023] Further, by calculating the initial access heat value of the spatial grid (the product of time density and spatial radius), the final heat value is obtained by diffusion and accumulation among adjacent grids, and the discretized step function is used to divide the block granularity level; the initial data blocks are generated by radially cutting with the grid center point, and the overlapping areas are re-divided with the center line to ensure that each data block belongs to only a single grid. This solution realizes the intelligent hierarchical cutting of data blocks based on heat diffusion and step function, taking into account the fine processing of hot spots and the resource saving of non-hot spots. By re-dividing the center line of the overlapping area, the ambiguity of data block ownership is avoided, and the accuracy of space segmentation and the clarity of data management are improved.

[0024] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0026] Figure 1 It shows a flowchart of a dynamic encryption method for geographic information data based on spatio-temporal features provided by the present application; Figure 2 It shows a schematic structural diagram of a dynamic encryption system for geographic information data based on spatio-temporal features provided by the present application; Figure 3 It shows a schematic structural diagram of a computing device provided by the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] In order to enable those skilled in the art to better understand the solution of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application.

[0028] In some of the processes described in the specification, claims, and the above-mentioned drawings of this application, a number of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The operation numbers, such as 101, 102, etc., are only used to distinguish different operations, and the numbers themselves do not represent any execution order. Additionally, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0029] Researchers have found that in the scenario of multi-institutional geographic information data sharing, traditional static encryption methods are difficult to adapt to the dynamically changing access requirements, and there are problems such as rigid key management, inaccurate permission control, and insufficient security for cross-institutional data aggregation. Based on this, a dynamic encryption method for geographic information data based on spatio-temporal characteristics is provided. This method can dynamically adjust the encryption strategy according to the spatio-temporal characteristics of user access behavior, and achieve fine-grained permission control and secure cross-institutional data aggregation.

[0030] The technical solution of this application is applicable to scenarios that require multi-institutional collaborative processing of geographic information data, such as smart cities, emergency management, and traffic planning, and is particularly applicable to application environments where the dynamic hotspots change, and high requirements for data security and sharing efficiency.

[0031] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0032] Figure 1 The flowchart of a dynamic encryption method for geographic information data based on spatio-temporal characteristics is provided for the embodiments of this application, as Figure 1 shown. This method includes: 101. Determine the dynamic hotspots according to the obtained timestamp data, geographic coordinate data, and access frequency data of user access requests in the multi-institutional sharing scenario; Timestamp data: Records the exact time of user access requests and is used to analyze the time distribution pattern.

[0033] Geographic coordinate data: The user access location (latitude and longitude), which supports spatial distribution analysis.

[0034] Access frequency data: The number of accesses to the same area within a unit time, which reflects the regional popularity.

[0035] Dynamic hot spot area: An access-intensive area that changes dynamically based on time, space, and frequency, with spatio-temporal locality characteristics.

[0036] In the embodiments of the present application, first, through the access log interface of the multi-institution sharing platform, timestamp data (such as 2023-10-01 08:30:00), geographic coordinate data (such as longitude 116.404, latitude 39.915), and access frequency data (such as 50 times in 5 minutes) of user requests are collected in real time. Secondly, the collected timestamp data, geographic coordinate data, and access frequency data are input into an improved spatio-temporal weighted DBSCAN clustering algorithm, where the time decay coefficient λ = 0.1 (decaying 10% per hour) and the spatial neighborhood radius ε = 500 meters are set, and density clustering is performed on the access points within a continuous 1-hour time window to obtain multiple clustering clusters. Then, the boundary vertex coordinates of each clustering cluster are calculated through the convex hull algorithm (such as generating a polygon vertex set {P1(116.402,39.914), P2(116.405,39.917)...}), and the total access frequency within the cluster is statistically calculated. Finally, the clustering area with the total access frequency exceeding the threshold (such as 200 times / hour) is marked as a dynamic hot spot area, and a hot spot area list containing the regional boundary coordinates and real-time density is output, providing a spatial analysis basis for the construction of the weight matrix in step 102.

[0037] The traffic management department of a certain city jointly with shared mobility platforms (such as Platform 1, Platform 2), based on the user cycling request data (timestamp, GPS coordinates, usage frequency) shared by multiple institutions, divides the spatio-temporal grid with a 30-minute time window and a 1-kilometer spatial radius, and counts the number of cycling times in each grid. High-frequency areas are screened out through an improved ST-DBSCAN algorithm: For example, during the morning rush hour (7:00 - 9:00) on weekdays, in the city center business district (longitude and latitude range E116.30° - 116.35°, N39.90° - 39.93°), if the cycling frequency exceeds 500 times / hour, the adjacent grids are merged to generate a dynamic hot spot area. This dynamic hot spot area is marked as the "morning rush hour core area" for subsequent resource scheduling optimization.

[0038] 102. Generate a multi-dimensional access weight matrix according to the time distribution density and spatial coverage radius of the dynamic hot spot area, where the multi-dimensional access weight matrix includes a time dimension decay factor and a spatial dimension distance factor; Time distribution density: The change rate of the access frequency of the hot spot area in different time periods, used to quantify time decay.

[0039] Spatial coverage radius: The geographical coverage range of the hot spot area, used to quantify spatial decay.

[0040] Time dimension decay factor: A time-weight decay coefficient based on an exponential function (such as e λt).

[0041] Space dimension distance factor: A space-weight decay coefficient based on a Gaussian kernel function (such as e d2 / σ).

[0042] In the embodiments of this application, the generation process of the multi-dimensional access weight matrix can be summarized as follows: First, receive dynamic hotspot area data, and calculate the time distribution density (such as the access volume proportion in the early peak period from 8:00 to 9:00 is 65%) and the spatial coverage radius (the minimum circumscribed circle radius R = 550 meters) for each hotspot area; Subsequently, construct a two-dimensional matrix framework of 24 (time slots) × 100 (spatial grids), calculate the time dimension decay factor (Wt = e^(-0.1·Δt), where Δt is the difference from the start time of the hotspot) and the space dimension distance factor (Ws = 1 - d / R, where d is the Euclidean distance from the grid to the hotspot centroid) for each grid cell respectively, and fuse the spatio-temporal decay effect through linear superposition (α = 0.6 time weight, β = 0.4 space weight) to generate an access heat value normalized to the [0, 1] interval; The finally output multi-dimensional access weight matrix provides a decision basis for subsequent data chunking through a two-dimensional quantization model that fuses exponential time decay and Gaussian space decay.

[0043] A dynamic hotspot area modeling method based on the decay of the time dimension decay factor and the space dimension distance factor is as follows: First, according to the dynamic hotspot data in the core area of the subway station during the morning rush hour, extract the time distribution density (the access volume proportion in the period from 8:00 to 9:00 reaches 65%) and the spatial coverage radius (1.2 kilometers); Then construct a multi-dimensional access weight matrix including the time dimension decay factor (an exponential function with λ = 0.15, such as at 9:00, Wt = e^(-0.15 1) ≈ 0.8607) and the space dimension distance factor (based on a Gaussian kernel function, such as when the distance from the core is 500 meters, Ws ≈ 0.2105); Through spatio-temporal two-factor coupling calculation (Wij = Wt × W_s), generate a 24×400 multi-dimensional access weight matrix (such as at 9:00, 500 meters from the core, Wij ≈ 0.1815), and normalize it to the [0, 1] interval. The multi-dimensional access weight matrix can accurately quantify the access priorities of different spatio-temporal units, where high-weight areas (Wij ≥ 0.3) correspond to core spatio-temporal hotspots, and low-weight areas (Wij < 0.1) reflect marginal decay areas, providing a quantitative decision basis for resource dynamic scheduling.

[0044] 103. Analyze the access heat value of the spatial grid in the multi-dimensional access weight matrix, determine the corresponding block granularity level according to the numerical range of the access heat value, and perform a spatial segmentation operation on the target geographic information data according to the block granularity level to generate multiple data blocks matching the spatial grid; Access heat value: The weighted sum of the element values in the grid weight matrix, reflecting the spatio-temporal comprehensive access intensity.

[0045] Block granularity level: Divide the data block size level according to the heat value (e.g., fine granularity in high heat areas, coarse granularity in low heat areas).

[0046] In the embodiment of the present application, first, analyze the multi-dimensional access weight matrix generated in step 102, and divide the access heat value of the spatial grid in the multi-dimensional access weight matrix into three levels according to a preset rule: high heat area (≥0.8), medium heat area (0.5 - 0.8), and low heat area (<0.5). Secondly, adopt a differentiated block strategy for different heat areas: use quadtree recursive segmentation for the high heat area to a precision of 100 meters (e.g., divide a 1 km² grid into 100 sub-grids of 100 m × 100 m), keep the original 1 km² grid for the medium heat area, and merge the low heat areas into large blocks of 5 km². Then, perform spatial segmentation on the target geographic information data (such as road network data) based on Geohash coding to generate multiple data blocks matching the spatial grid (such as all road data within the longitude range of 116.400 - 116.450 and latitude range of 39.900 - 39.950 corresponding to block B0231). Finally, establish an R-tree spatial index to accelerate data retrieval, and output the segmented structured geographic data set, providing data input for the dynamic key generation in step 104.

[0047] Calculate the access heat value Hij of each grid according to the multi-dimensional access weight matrix, divide the central grid of the commercial area (heat value > 1.0) into fine-grained blocks of 50 m × 50 m, divide the peripheral area (heat value 0.6 - 1.0) into medium-grained blocks of 100 m × 100 m, and merge the remote areas (heat value < 0.6) into coarse-grained blocks of 200 m × 200 m. Use the quadtree algorithm to recursively segment the city map to ensure that the data blocks in high heat areas (such as around subway stations) are more refined, supporting the accuracy of subsequent encryption and query.

[0048] 104. Generate a dynamic key seed based on the time dimension decay factor and the space dimension distance factor; Dynamic key seed: A random number combining the time decay factor and the space distance factor, serving as the basis for key generation.

[0049] In the embodiment of the present application, first, the spatial dimension distance factor of the target space grid is extracted from the multi-dimensional access weight matrix in step 102 (for example, the heat value 0.82 corresponds to W_t = 0.85, W_s = 0.78), which is quantized into 16-bit fixed-point numbers (0x85C0, 0x78F4) and binary concatenated (0x85C078F4). Secondly, using the pre-institutional shared key K_shared as the key for the HMAC-SHA256 algorithm, the concatenated spatio-temporal parameters are hashed to generate an initial seed value (such as 0x5a3d...f209). Then, random perturbations are added to the seed value through the Gaussian noise injection module (μ = 0, σ = 0.01) to enhance the anti-cracking ability. Finally, a 128-bit dynamic key seed is output (such as Dynamic_Seed = 0x5a3d...f20a), which will be used as the basis for generating the temporary encryption key in step 105.

[0050] Encode the time dimension decay factor λ = 0.15 and the spatial radius R = 1200 meters as the string "0.15_1200", generate the initial seed through SHA-256 hashing, and superimpose the microsecond noise of the current system time (such as 2025-04-09 08:30:45.678) to generate the dynamic key seed. For example, the hashing result is "a1b2c3...", and after superimposing the noise, it becomes "a1b2c3...d4e5f6", enhancing the unpredictability of the key.

[0051] 105. Perform a discrete logarithm operation on the dynamic key seed to generate a temporary encryption key, use the temporary encryption key to perform a homomorphic encryption operation on the multiple data blocks, and embed a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix in the encrypted data blocks; Discrete logarithm operation: Generate a temporary key based on k = H(s)modp of the elliptic curve (ECC).

[0052] Spatio-temporal feature identifier: Metadata embedded in the ciphertext, including permission level, timestamp, location hash. [[ID=?]]

[0053] In the embodiment of the present application, first, perform elliptic curve discrete logarithm operation (selecting the secp256k1 curve) on the dynamic key seed output by step 104 to generate a temporary public key PK (0x03a5...c7e1) and a private key SK (0x8d2...f4b). Secondly, adopt the Paillier homomorphic encryption algorithm to encrypt the multiple data blocks output by step 103 using PK (for example, encrypt the 2MB road data of block B0231 to generate a 3.2MB ciphertext C_B0231). Then, encrypt the time dimension decay factor and the space dimension distance factor of the corresponding grid in the multi-dimensional access weight matrix (such as the timestamp "20231001T0800", grid ID "G32", permission level "L2") through AES-128 to generate a 128-bit spatio-temporal feature identifier. Finally, use the LSB steganography technique to embed the identifier into the check bit at the end of the ciphertext, and output the encrypted data block with the spatio-temporal feature identifier (such as C_B0231'), ensuring that the data source and access permission can be verified during the cross-institutional data aggregation in the subsequent step 106.

[0054] Perform discrete logarithm operation on the dynamic seed using the elliptic curve secp256k1 to generate a temporary private key k = 0x5A3D and a public key K = kG. Encrypt the data block using the Paillier homomorphic encryption algorithm and embed the spatio-temporal feature identifier in the ciphertext, such as "L2_20250409_116.32E39.91N", where L2 represents the permission level (only allowing access by the traffic management department), followed by the timestamp and the hash value of the central coordinates. The encrypted data block can quickly match the permission and area through the identifier.

[0055] 106. When receiving a cross-institutional data aggregation request, extract the permission level information in the spatio-temporal feature identifier from the encrypted data block, perform ciphertext superposition operation on multiple encrypted data blocks based on the homomorphic encryption property, and at the same time verify the matching degree between the user identity identifier of the requester and the permission level information to complete the secure aggregation operation.

[0056] Permission level information: The access permission defined in the identifier (such as L2 only allowing access by the traffic management department).

[0057] Ciphertext superposition: Combine multiple encrypted data blocks using the homomorphic additivity of Paillier.

[0058] In the embodiments of the present application, first, a cross-institutional data aggregation request is received, which includes a target area range (such as longitude 116.40 - 116.45, latitude 39.90 - 39.95) and a user identity certificate (such as the digital certificate of Institution B). Secondly, extract the 128-bit spatio-temporal feature identifier hidden in the LSB from the encrypted data blocks generated in step 105, and obtain the original spatio-temporal feature data (such as timestamp "20231001T0800", grid ID "G32", permission level "L2") after decrypting with AES-128.

[0059] Next, perform triple verification: Timeliness verification: Check whether the request time is within the validity period of the identifier timestamp (such as ±2 hours); Spatial permission verification: Compare the inclusion relationship between the request area and the identifier grid ID (such as the request area completely contains grid G32); Institutional permission verification: Verify whether the institutional ID in the requester's certificate matches the identifier permission level through the zero-knowledge proof protocol (such as verify whether Institution B has L2-level permission).

[0060] After the verification passes, perform the Paillier homomorphic addition operation on all encrypted data blocks (such as C_B0231', C_B0232', etc.) within the target area to generate the aggregated ciphertext C_sum. Finally, return the aggregated ciphertext C_sum and the new spatio-temporal feature identifier (integrating the permissions of each data block) to the requester. The requester can obtain the aggregated data after decrypting with the temporary private key SK generated in step 105, and at the same time, the system automatically destroys the temporary key pair of this session. The whole process completes secure aggregation without decrypting the original data and ensures that the data usage complies with spatio-temporal permission constraints.

[0061] When the environmental protection department applies for aggregating morning rush hour cycling data, the system extracts the identifier "L2" from the ciphertext to verify its permission (which needs to match the institutional attribute in the digital certificate), and then performs the homomorphic addition operation on 10 encrypted data blocks to obtain the aggregated ciphertext of the total number of cycling times. After decryption, it shows that the total morning rush hour cycling volume is 12,345 times, which is used to optimize the shared bicycle scheduling. At the same time, the blockchain records the aggregation operation log (such as "2025-04-09 09:15:00 Environmental Protection Department - L2 - Aggregation Successful"), realizing the full-process audit.

[0062] In some embodiments, parsing the access heat value of the spatial grid in the multi-dimensional access weight matrix, determining the corresponding block granularity level according to the numerical range of the access heat value, and performing a spatial segmentation operation on the target geographic information data according to the block granularity level to generate multiple data blocks matching the spatial grid, including: 201. Mark the initial access heat value within the coverage range for each spatial grid in the multi-dimensional access weight matrix. The initial access heat value is the product of the time distribution density and the spatial coverage radius. Multi-dimensional access weight matrix: A set of spatial grids composed of three dimensions: longitude, latitude, and time. Each grid records data access characteristics.

[0063] Time distribution density: The statistical value of the number of times the grid is accessed per unit time (such as the access volume per hour).

[0064] Spatial coverage radius: The maximum geographical distance from the grid center point to its boundary (such as 500 meters).

[0065] Initial access heat value: A comprehensive index of the grid access frequency and spatial influence. The calculation formula is: heat = time density × coverage radius.

[0066] In the embodiments of the present application, first, obtain the historical access log, and calculate the time distribution density of each grid through time series analysis (such as sliding window counting). At the same time, calculate the spatial coverage radius of the spatial grid based on the geographical coordinates (using the spherical distance formula). Multiply the time distribution density and the spatial coverage radius to generate the initial access heat value. For example, the initial access heat value of a 10km×10km grid during the peak period (time density = 5 times / minute) is 5×10 = 50.

[0067] 202. Construct a heat diffusion window based on the adjacent relationship between spatial grids, and perform horizontal propagation and accumulation of the initial access heat value within the heat diffusion window to obtain the access heat value. Adjacent relationship: Adopt the eight-neighborhood topological structure (including the grids above, below, left, right, and diagonal).

[0068] Heat diffusion window: A 3×3 convolution kernel, with the central weight being 0.5 and the adjacent unit weights being 0.2.

[0069] Horizontal propagation and accumulation: Realize the weighted superposition of heat values through spatial convolution.

[0070] In the embodiments of the present application, first construct a heat diffusion window K = [[0.2, 0.2, 0.2], [0.2, 0.5, 0.2], [0.2, 0.2, 0.2]]. Perform a two-dimensional convolution operation on the initial access heat value using the heat diffusion window (adopting zero-padding boundary processing). The new access heat value H' of each spatial grid is H' = Σ(K(i,j) × H_neighbor). For example, if the initial access heat of a central spatial grid is 50 and the average initial access heat of its eight neighbors is 30, then the access heat value after diffusion = 50×0.5 + 30×0.2×8 = 25 + 48 = 73. This process is iteratively executed 3 times to finally generate the global access heat value.

[0071] 203. Input the access heat value into a preset discretization step function to divide the corresponding block granularity levels according to each step interval through the discretization step function; Discretization step function: A non-linear mapping function based on heat percentiles.

[0072] Block granularity levels: L1 - L5 levels. L1 is the coarsest granularity (allowing 1 block), and L5 is the finest granularity (allowing 16 blocks).

[0073] In the embodiments of the present application, first, a distribution histogram of the global access heat value is statistically calculated, and the threshold of the step interval is determined by inputting it into a preset discretization step function. For example, the access heat value is divided into 5 step intervals: L1 level (0 - 20% percentile): ≤100, L2 level (20 - 40% percentile): 101 - 200... and so on. Each grid determines the corresponding block granularity level according to the step interval into which its access heat value falls. For example, if the access heat value of a spatial grid is 250, it corresponds to the L3 level (allowing a maximum of 4 blocks). The block granularity level will control the cutting density of subsequent block operations.

[0074] 204. Determine the maximum number of blocks allowed within the spatial grid according to the block granularity level, and perform radial cutting at equal angular intervals along the longitude and latitude directions with the center point of the spatial grid as the reference to form multiple initial data blocks; Radial cutting: Using the grid center as the origin, generating longitude cuts at angular intervals.

[0075] Equal angular interval: Δθ = 360° / (N×2), where N is the maximum number of blocks.

[0076] In the embodiments of the present application, first, query the block quantity look-up table according to the block granularity level (for example, L3 level corresponds to N = 4). Using the center point (λ0, φ0) of the spatial grid as the reference, generate a cutting meridian every Δθ = 360 / (4×2) = 45°. For example, for the L3 level spatial grid, generate a total of 8 cutting lines at 0°, 45°, 90°... 315°, dividing the spatial grid into 8 sectors. The boundary line equation of each sector adopts the azimuth angle calculation formula in the spherical coordinate system: φ = φ0 + tanθ×(λ - λ0). In this process, multiple initial data blocks are generated, and each block is defined by its azimuth angle range.

[0077] 205. When the multiple initial data blocks generate an overlapping area with the cutting lines of adjacent spatial grids, re-divide the boundaries of each initial data block with the center line of the overlapping area as the boundary to generate multiple data blocks, and each data block belongs to a single spatial grid. '

[0078] Overlapping area: The overlapping geographical area formed by adjacent grid cutting lines at the boundary.

[0079] Centerline redivision: Adjust the block boundary with the midline of the overlapping area as the boundary.

[0080] In the embodiment of the present application, first, the intersection points of adjacent spatial grid cutting lines are detected through a spatial index (R-tree). For each overlapping area, the intersection point set of the equations of two adjacent cutting lines is calculated, and the perpendicular bisector is generated as the new boundary. For example, if the 45° cutting line of spatial grid A intersects the 30° cutting line of spatial grid B at point P, then with point P as the center, the boundary of each initial data block is redetermined along the centerline of the included angle between the two lines to generate multiple data blocks. Finally, it is ensured that each data block belongs to only a single spatial grid, and the boundary line meets the topological closure requirements.

[0081] The following is a specific embodiment: A certain intelligent warehouse divides the site into grids of 100m×100m. The access frequency of the forklift in the A23 spatial grid is counted as 12 times / minute (time density), its coverage radius is 100m, and the initial heat = 12×100 = 1200. After diffusion with a 3×3 Gaussian kernel, the access heat of the A23 spatial grid rises to 1800, corresponding to L4 level (allowing 8 sub-blocks). The A23 spatial grid is cut into 8 sectors at 45° intervals with the center of the A23 spatial grid as the center. The cutting line of the adjacent spatial grid B15 overlaps with the 135° line of the A23 spatial grid at the warehouse passage. By calculating the midline of the overlapping area, the original sector boundary is adjusted to be divided along the centerline of the passage, and finally 8 independent shelf management blocks are generated.

[0082] This solution realizes the adaptive sub-blocking of multi-dimensional spatial data through dynamic calculation of heat values, spatial diffusion conduction, discrete sub-block control, and boundary optimization processing. Compared with the fixed sub-blocking method, it improves the sub-blocking accuracy of the access hot spot area, reduces the overlapping rate of data blocks across spatial grids, and dynamically adjusts the sub-blocking structure according to the real-time access pattern. This method is particularly suitable for scenarios such as logistics scheduling and meteorological grid computing that need to take into account spatial proximity and access heat.

[0083] In some embodiments, based on the time dimension decay factor and the spatial dimension distance factor, a dynamic key seed is generated, including: 301. Perform timestamp difference operation based on the time dimension decay factor to generate a time decay coefficient; Time dimension decay factor: A dynamic parameter representing the validity period of the key, which is negatively correlated with the time difference.

[0084] Timestamp difference operation: Calculate the difference between the current time and the key reference time (unit: second).

[0085] Time decay coefficient: The time weight value processed by the exponential decay function.

[0086] In the embodiment of the present application, first, a time - stamp difference operation is performed to calculate the time difference Δt (unit: second) between the current time - stamp (T_now) and the preset key reference time (T0); subsequently, Δt is input into an exponential decay function constructed based on a time - dimension decay factor (the mathematical form is α = e^(-λ·Δt), where λ is the preset decay rate) to generate a time - decay coefficient α, and this coefficient decreases exponentially as Δt increases; finally, the validity of the key is dynamically determined according to the comparison result between the time - decay coefficient α and a preset threshold, realizing a protection mechanism in which the security strength of the key automatically decays over time.

[0087] 302. Perform a geographic - coordinate difference operation based on the spatial - dimension distance factor to generate a spatial - distance coefficient; Spatial - dimension distance factor: A dynamic parameter representing the credibility of the device location, which is negatively correlated with the coordinate deviation.

[0088] Geographic - coordinate difference operation: Calculate the spherical distance between the current location and the registered location.

[0089] Spatial - distance coefficient: A normalized spatial - weight value.

[0090] In the embodiment of the present application, first, a geographic - coordinate difference operation is performed to calculate the actual distance d between the current location and the registered location based on a spherical - distance formula (such as the Haversine formula); subsequently, the actual distance d is input into a normalization function constructed by the spatial - dimension distance factor (such as β = 1 / [1+(d / D_max)^k], where D_max is the maximum allowable deviation distance and k is the curvature factor) to generate a spatial - distance coefficient β; the spatial - distance coefficient decreases non - linearly as the actual distance d increases (for example, when d = 0, β = 1; when d = D_max, β≈0.5), thereby dynamically mapping the geographic deviation to a security - weight parameter in the spatial dimension and jointly constraining the key generation with the time - decay coefficient.

[0091] 303. Input the time - decay coefficient and the spatial - distance coefficient into a non - linear transformation function to generate a dynamic key seed.

[0092] Non - linear transformation function: An irreversible calculation model.

[0093] Dynamic key seed: A 128 - bit random number used to derive an encryption key.

[0094] In the embodiments of the present application, the time decay coefficient α and the spatial distance coefficient β are spliced into an input data stream, which is input into a constructed non-linear transformation function. Through iterative calculation and post-processing (such as taking the binary mantissa and performing a hash operation), an irreversible dynamic key seed (a 128-bit random number) is generated. The final output result has the dynamic constraint characteristics of both time and space dimensions, ensuring that the key cannot be cracked through reverse derivation or parameter tampering.

[0095] The following is a specific embodiment: In the intelligent lock system of shared bicycles, the dynamic key generation scheme realizes security authentication through the following complete process: The system first records the initial timestamp (T0 = 2024-03-10 12:00:00) and the registration location (the center point of Beijing). When the user requests to unlock the bicycle within 500 meters (β≈0.995) after 30 minutes (T1 = 12:30:00), the time decay coefficient (α≈0.368) and the spatial distance coefficient (β) are calculated, and the key seed (S = 0.712) is generated through chaotic mapping mixing. Finally, an AES-256 key is derived; if the bicycle is moved 10 kilometers away (β→0) or exceeds the time limit by 1 hour (α→0), the key will immediately become invalid and trigger the lock protection.

[0096] Through the dynamic coupling of the time decay coefficient and the spatial distance coefficient, this scheme realizes the real-time adaptive update of the key: The time dimension enforces periodic invalidation, and the spatial dimension prevents out-of-bounds use. Combining non-linear transformation ensures that the key seed is unpredictable. In the scenario of shared bicycles, it effectively defends against key cracking, device cloning, and cross-regional theft, reduces the key change response time, increases the key invalidation rate in displacement or overtime scenarios, and significantly improves the security of Internet of Things devices.

[0097] In some embodiments, a discrete logarithm operation is performed on the dynamic key seed to generate a temporary encryption key, and the temporary encryption key is used to perform a homomorphic encryption operation on the multiple data blocks, and a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix is embedded in the encrypted data block, including: 401. Perform a cyclic shift operation on the initial value of the dynamic key seed, perform an exclusive OR operation on each element corresponding to the initial value and the adjacent element after shifting, generate an extended key sequence, and select elements that meet the preset parity condition according to the parity distribution characteristics of the elements in the extended key sequence to generate a temporary encryption key; Dynamic key seed: A random number sequence generated by the system, used as the encryption basic parameter.

[0098] Cyclic shift operation: After connecting the seed sequence end to end to form a ring structure, perform displacement calculation.

[0099] Parity distribution feature: The statistical result of the parity of the number of binary 1s in the extended key sequence.

[0100] In the embodiment of the present application, first, a cyclic shift operation is performed on the 32-bit dynamic key seed, shifted 3 bits to the left, and each original element is exclusive-ORed with the corresponding element after shifting bit by bit to generate a 64-bit extended key sequence; subsequently, the parity distribution feature of each byte is statistically analyzed, the bytes containing an odd number of 1s are screened out, and a 128-bit temporary encryption key is generated through SHA-256 hash compression. This process enhances the randomness of the key through displacement diffusion and feature screening.

[0101] 402. Divide the temporary encryption key into multiple sub-key segments, each sub-key segment corresponding to the encryption position of a data block, calculate the starting offset of the sub-key segment in the temporary encryption key sequence according to the spatial grid coordinates of the data block, and intercept a sub-key segment of the corresponding length according to the offset as the encryption parameter of the data block to perform a homomorphic encryption operation on the multiple data blocks to generate encrypted data blocks; Sub-key segment: An encryption segment formed by equally dividing the temporary key according to the number of data blocks.

[0102] Spatial grid coordinates: The (x, y) position identifier of the data block in the two-dimensional matrix.

[0103] Starting offset: The starting position of the sub-key interception calculated according to the coordinates.

[0104] In the embodiment of the present application, the 128-bit temporary encryption key is divided into 16 8-bit sub-key segments. The offset calculation formula for each data block is offset = (x × y mod 8) × 16, where (x, y) are the spatial grid coordinates; 16 consecutive bits are intercepted according to the starting offset as the encryption parameter, and the Paillier homomorphic algorithm is used to encrypt the data block to generate an encrypted data block, maintaining the computable property of the ciphertext. This step realizes the dynamic binding of the key and the spatial position.

[0105] 403. Bit-cross weave the time distribution density feature code and the spatial coverage radius feature code of the spatial grid from the multi-dimensional access weight matrix to generate a spatio-temporal feature identifier, and embed the spatio-temporal feature identifier into the corresponding encrypted data block.

[0106] Spatio-temporal feature identifier: A 128-bit digital fingerprint that fuses time and space features.

[0107] Bit-cross weave: A coding method that alternately combines two feature codes according to odd and even bits.

[0108] In the embodiments of the present application, a time distribution density feature code (64 bits) and a space coverage radius code (64 bits) are extracted from a multi-dimensional access weight matrix, and a spatio-temporal feature identifier is generated by taking the time code for odd bits and the space code for even bits; the spatio-temporal feature identifier is embedded into the redundant bits of an encrypted data block through the LSB steganography technique to form an encrypted data block with spatio-temporal markings. This process realizes the feature association between the encrypted data and the business scenario.

[0109] The following is a specific embodiment: Generate a dynamic key seed 0x3A7F based on the real-time vehicle-mounted GPS data, and obtain an extended sequence 0xB2E4... after cyclic shift operation and exclusive OR operation, and screen the parity distribution characteristic bytes to generate a temporary encryption key. Divide the traffic flow data into a 4×4 space grid, calculate the starting offset 18 for the data block at the (2,3) position, and intercept the sub-key 0x5C... to perform homomorphic encryption. Integrate the morning peak time distribution density feature code (08:00 code) and the business district space coverage radius feature code (500m code) to generate a spatio-temporal feature identifier 0xAE... and embed it into the encrypted data block.

[0110] This solution ensures the basic security of encryption through the generation of a dynamic key seed, realizes fine-grained access control through space-adaptive key distribution, and finally constructs a traceable encrypted data system through the integration of spatio-temporal feature identifiers. The three steps are connected in sequence to form a closed loop: the output temporary key is used for data block encryption, and then the spatio-temporal feature marking is completed using the spatial grid coordinate information before processing. In spatio-temporally sensitive scenarios such as smart cities, it can not only prevent the risk of key leakage but also support the spatio-temporal feature retrieval of encrypted data, overall improving data security and business compliance.

[0111] In some embodiments, when a cross-institutional data aggregation request is received, extract the permission level information in the spatio-temporal feature identifier, perform ciphertext superposition operation on multiple encrypted data blocks based on the homomorphic encryption property, and at the same time verify the matching degree between the user identity identifier of the requestor and the permission level information to complete the secure aggregation operation, including: 501. When a cross-institutional data aggregation request is received, extract the spatio-temporal feature identifier from the encrypted data block, and extract the permission level information composed of a digital sequence from a fixed position of the spatio-temporal feature identifier, compare the permission level information item by item with a pre-stored institutional permission mapping table, and screen out multiple target encrypted data blocks that are allowed to participate in the superposition operation from multiple encrypted data blocks; Spatio-temporal feature identifier: A 128-bit feature code embedded in the encrypted data block, containing spatio-temporal attributes and permission information.

[0112] Permission level information: A 4-bit digital sequence of data access permissions, pre-buried in bits 32-35 of the identifier.

[0113] Institutional Permission Mapping Table: A table recording the correspondence between institutional IDs and permitted access permission levels In the embodiments of the present application, first, the LSB steganographic bits of the encrypted data block are parsed to extract the spatio-temporal feature identifier, and the permission level information at a fixed position is intercepted; then, the permission range of the requesting institution of the cross-institutional data aggregation request in the pre-stored institutional permission mapping table is queried, and multiple target encrypted data blocks with permission level information within the permitted range are filtered out. For example, if institution A has permission level information of "1100", only the encrypted data blocks whose first two bits of the permission level information code are "11" are selected.

[0114] 502. Split the user identity identifier of the requester carried in the cross-institutional data aggregation request into a header verification code and a tail permission identifier. The header verification code is converted into a binary mask through a preset permutation rule and performs bitwise overlay matching with the verification area of the spatio-temporal feature identifier; the tail permission identifier reversely deduces the theoretical identity feature identifier through the permission level information. When the number of consecutive matching times between the theoretical identity feature identifier and the verification bit of the user identity identifier reaches a preset threshold, the superimposed operation channel is activated; User Identity Identifier: A 64-bit identity credential, including a 16-bit header verification code and a 48-bit tail permission identifier.

[0115] Binary Mask: A 16-bit mask (such as 0xF0A3) converted from the header verification code through a permutation table.

[0116] Theoretical Identity Feature Identifier: A standard identity feature deduced inversely according to the permission level.

[0117] In the embodiments of the present application, the user identity identifier is split into a header verification code and a tail permission identifier. The header verification code generates a binary mask through a preset permutation rule (such as parity bit swapping) and performs an AND operation with the spatio-temporal identifier verification area to verify the matching degree; the tail permission identifier reversely deduces the theoretical identity feature identifier through the permission level information, and the channel is activated when 8 consecutive verification bits match. For example, the permission "1101" corresponds to the theoretical identity feature identifier "1011_xxxx", and the corresponding bits of the identifier need to be matched.

[0118] 503. Individually split the multiple target encrypted data blocks into multiple ciphertext segments, and input the ciphertext segments with the same position index in the multiple encrypted data blocks into the superimposed operation channel. Generate a dynamic weight factor based on the permission level information corresponding to the multiple encrypted data blocks, perform weighted multiplication and accumulation on the ciphertext segments within the same channel, and generate the ciphertext segment corresponding to the position in the aggregated ciphertext; Ciphertext Segment: Evenly divide a 256-bit encrypted data block into 8 32-bit segments.

[0119] Dynamic weight factor: A multiplier for permission level calculation (e.g., level 3 corresponds to a weight of 0.7).

[0120] Product accumulation: A homomorphic operation of adding weighted ciphertext segments at the same position.

[0121] In the embodiments of the present application, after each target encrypted data block is split into ciphertext segments, dynamic weight factors are assigned according to the permission level information (e.g., levels 1-4 correspond to 0.3 / 0.5 / 0.7 / 1.0 respectively). Under the Paillier homomorphic encryption system, the product accumulation of the ciphertext segments with the dynamic weight factors is performed to generate the ciphertext segments at the corresponding positions in the aggregated ciphertext. Target encrypted data with higher permissions has a greater weight to affect the result.

[0122] 504. During the superposition process, the number of ciphertext segments that have been superimposed for each encrypted data block in the aggregated ciphertext is statistically calculated in real time. If the superimposition quantity of any target encrypted data block exceeds the maximum superimposed segment number threshold corresponding to the permission level information of the target encrypted data block, the superimposition operation channel corresponding to the target encrypted data block is closed.

[0123] Maximum superimposed segment number threshold: The maximum number of data segments allowed to be contributed by the permission level.

[0124] Channel closing: Terminate the aggregation operation of the subsequent segments of this data block.

[0125] In the embodiments of the present application, the number of ciphertext segments that have been superimposed for each encrypted data block in the aggregated ciphertext is statistically calculated in real time. When the contribution number of a target encrypted data block (e.g., permission level 2) reaches the threshold (e.g., 3 segments) corresponding to the permission level information, the transmission of its remaining ciphertext segments is immediately terminated. Dynamic fusing is achieved through a counter to prevent low-permission target encrypted data from overly affecting the result.

[0126] The following is a specific embodiment: When Hospital A with a permission level information of 3 initiates a cross-institutional data aggregation request, the system first screens out the CT image data with a permission level information code of "11xx" from the encrypted medical record data blocks; subsequently, the user identity identifier 0xE53D... provided by Hospital A is verified. Through the binary mask matching of the header verification code and the reverse derivation of the tail permission identifier, the superimposition operation channels of 3 eligible target data blocks are successfully activated; after splitting 8 CT data blocks into ciphertext segments, homomorphic encryption product accumulation is performed according to the weight of 0.7 corresponding to the permission level information of 3 for Hospital A; during the aggregation process, the contribution of each data block is monitored in real time. When a target encrypted data block has reached the maximum contribution threshold of 5 segments allowed by the permission level information of 3, the superimposition operation channel of this data block is immediately closed. This example fully demonstrates the entire process from permission verification, identity authentication to dynamic weighted aggregation, realizing the safe, compliant sharing and analysis of medical data while ensuring patient privacy.

[0127] This solution ensures data access compliance through permission grading screening and two-factor authentication, and achieves fine-grained data fusion by using dynamic weighted aggregation and contribution volume control. In scenarios such as medical joint analysis, it not only protects patient privacy (low-permission institutions can only obtain limited data), but also maintains the value of data analysis (high-permission institutions can obtain more complete aggregation results). At the same time, homomorphic encryption technology is used to ensure that data is not decrypted throughout the process. Four steps form a closed-loop control to establish a secure access mechanism, perform controlled aggregation, and ultimately achieve the secure compliance and effective utilization of cross-institutional data.

[0128] In some embodiments, performing a timestamp difference operation based on the time dimension decay factor to generate a time decay coefficient includes: 601. Record the first marked timestamp value at the start time of the acquisition event and the second marked timestamp value at the current time of the event using the same time measurement rule; Time measurement rule: It refers to using standardized time representation methods such as UTC timestamp and Unix timestamp to ensure consistent time measurement units (such as millisecond-level precision).

[0129] First marked timestamp: The initial time record when the event is triggered (such as the moment when the user click action occurs).

[0130] Second marked timestamp: The current system time or the event processing time (such as the system time during data analysis).

[0131] In the embodiments of the present application, the original timestamp when the event is triggered is obtained through the system clock service (such as calling System.currentTimeMillis() in Java), and it is stored as the first marked timestamp value; when calculating the decay coefficient, the same clock service is called again to obtain the second marked timestamp value. The first marked timestamp value and the second marked timestamp value need to be processed by time zone standardization (such as converting to UTC+0 time zone) to ensure the accuracy of subsequent difference calculations. The first marked timestamp value serves as the reference anchor point, and the second marked timestamp value serves as the dynamic variable, jointly constituting the basic input for the time difference calculation.

[0132] 602. Perform a timestamp difference operation on the first marked timestamp value and the second marked timestamp value to obtain an initial time difference, and divide the initial time difference by the corresponding value of the preset basic time unit to obtain a standard time interval; Initial time difference: The original arithmetic difference between two timestamps (the unit is the same as the timestamp precision, such as a millisecond difference of 3600000).

[0133] Basic time unit: A standardized unit set according to business requirements (such as 1 hour = 3600000 milliseconds).

[0134] Standard time interval quantity: A dimensionless standardized time span (e.g., initial difference 3600000 ms / 3600000 ms = 1.0 time unit).

[0135] In the embodiment of the present application, first, an arithmetic subtraction operation is performed on the first marked timestamp value and the second marked timestamp value (second marked timestamp value - first marked timestamp value) to obtain an initial difference in milliseconds. Then, the initial difference is divided by a preset unit duration (such as the number of milliseconds corresponding to 1 hour) through a divider, and fixed-point arithmetic is used in this process to ensure accuracy. For example, when the initial difference is 3.6e6 milliseconds, dividing by 3.6e6 gives a standard time interval quantity of 1.0. This standardization process enables unified comparison of time differences at different granularities (such as second level / minute level).

[0136] 603. Extract a predefined time dimension decay factor, use the time dimension decay factor as the base number, and use the standard time interval quantity as the exponent, and perform an exponentiation operation of the base number to generate a time decay coefficient.

[0137] Time dimension decay factor: A predefined decay rate parameter (e.g., 0.5 means a 50% decay per time unit).

[0138] Exponentiation operation: Mathematical power function calculation (e.g., 0.5^1.0 = 0.5).

[0139] In the embodiment of the present application, a preset time dimension decay factor λ (usually a floating point number between 0 and 1) is read from the system configuration library and used as the base number of the power operation. The initial time difference quantity is divided by the corresponding value of the preset basic time unit to obtain the standard time interval quantity; using the standard time interval quantity t as the exponent, calculate λ^t through a math library (such as the pow function in C++). When t = 1.5 and λ = 0.6, the calculation result 0.6^1.5 ≈ 0.465 is the time decay coefficient. The time decay coefficient will decay exponentially as the time interval increases.

[0140] The following is a specific embodiment: In the scenario of e-commerce user behavior analysis, when user A clicks on a product at UTC time 2024-03-01 02:00:00 (timestamp 1709287200000 milliseconds), the system records this moment as the first marked timestamp; when the system calculates the decay coefficient at UTC time 2024-03-01 04:30:00 (timestamp 1709296200000 milliseconds), it first calculates the initial time difference of 9,000,000 milliseconds (2.5 hours), then divides it by the basic time unit of 3,600,000 milliseconds (1 hour) to obtain the standard time interval quantity of 2.5, and finally performs an exponential operation with the predefined decay factor of 0.7 as the base to generate a time decay coefficient of 0.7^2.5 ≈ 0.379. This coefficient decays the original weight of the user's click behavior to 37.9%, enabling the recommendation algorithm to dynamically reduce the impact of historical behavior and more accurately reflect the recent changes in user interests.

[0141] This solution realizes the dynamic adjustment of behavior weights through a standardized time difference and an exponential decay model. In the e-commerce scenario, the weights of click behaviors before the user's initial time are decayed to ensure that the recommendation system gives priority to recent behaviors (e.g., clicks with shorter time differences retain higher weights). This method solves the problem that traditional fixed weights cannot reflect timeliness, and at the same time, through UTC time zone unification and millisecond-level precision calculation, it avoids errors caused by cross-time zone or unit confusion, significantly improving the recommendation accuracy.

[0142] In some embodiments, bit-cross weaving the time distribution density feature code and the spatial coverage radius feature code of the spatial grid from the multi-dimensional access weight matrix to generate a spatio-temporal feature identifier, and embedding the spatio-temporal feature identifier into the corresponding encrypted data block, includes: 701. Extract the time distribution density feature code from the first predefined column of the multi-dimensional access weight matrix, and extract the spatial coverage radius feature code from the second predefined row; Multi-dimensional access weight matrix: A matrix composed of statistical features such as time access frequency and spatial position coordinates, where the first predefined column stores time distribution density features (such as the statistical distribution of time intervals), and the second predefined row stores spatial coverage radius features (such as the maximum coverage range of access locations).

[0143] Time distribution density feature code: An encoding that reflects the concentration degree of data access time, usually generated by entropy encoding or quantization encoding methods.

[0144] Spatial coverage radius feature code: An encoding that describes the spatial range of data access, usually generated by polar coordinate conversion or quantization after grid division.

[0145] In the embodiments of the present application, first, the time distribution density feature is extracted from the first column of the multi-dimensional access weight matrix, and Huffman coding or arithmetic coding is used for compression to generate a time distribution density feature code. At the same time, the spatial coverage radius feature is extracted from the second row, and the spatial coverage radius feature code is generated through polar coordinate conversion or grid quantization. Finally, the time density feature code and the spatial coverage radius feature code are used as a numerical array to prepare for further processing.

[0146] 702. Convert the time distribution density feature code and the spatial coverage radius feature code into fixed-length binary sequences respectively, and splice each bit of the two binary sequences crosswise in turn according to the alternating arrangement rule to generate a spatio-temporal feature identifier. Fixed-length binary sequence: The variable-length feature code is unified into a fixed number of bits (such as 16 bits or 32 bits) by zero-padding or truncation.

[0147] Alternating arrangement rule: The bit-by-bit crosswise splicing strategy, for example, bit0 of the time code, bit0 of the space code, bit1 of the time code, bit1 of the space code, and so on, until a complete identifier is generated.

[0148] In the embodiments of the present application, first, the time distribution density feature code and the spatial coverage radius feature code are respectively converted into fixed-length binary sequences. If the length of the binary sequence is insufficient, zeros are padded at the high positions. Subsequently, according to the alternating arrangement rule, bit operations (such as shifting and bitwise OR operations) are used to splice the converted binary sequences crosswise bit by bit, and finally a compact and unique spatio-temporal feature identifier is generated for the next step of embedding into the encrypted data block.

[0149] 703. Reserve a fixed-length embedding area at the head of the encrypted data block, and fill the spatio-temporal feature identifier into the embedding area in bit order, covering the values of the original data bits.

[0150] Embedding area: A fixed-length storage area reserved at the head of the encrypted data block for storing the spatio-temporal feature identifier, usually 32 bits or 64 bits.

[0151] Cover the values of the original data bits: Adopt the in-place replacement strategy, and the original data is ensured to be recoverable through a backup or verification mechanism.

[0152] In the embodiments of the present application, first, a fixed-length embedding area (such as 32 bits) is pre-computed and reserved at the head of the encrypted data block. The embedding area is implemented by modifying the padding strategy of the encryption algorithm or expanding the data block structure. Then, the spatio-temporal feature identifier generated in the previous step is written into this area bit by bit. An atomic write operation is used to ensure data consistency, and at the same time, the recoverability of the original data is retained through XOR operation. Finally, a lightweight verification algorithm (such as CRC8) is used to verify the embedding correctness, and the verification result is stored in the metadata area at the tail of the encrypted data block to form a complete embedding verification mechanism. This process is executed as a preprocessing step in the data encryption process to ensure the seamless integration of the spatio-temporal feature identifier and the encrypted data block.

[0153] The following is a specific embodiment: In the smart city video surveillance system, in the actual application of this solution, first, the system extracts the spatio-temporal feature data of a specific camera from the multi-dimensional access weight matrix. For example, the "1101" code is obtained from the 3rd column, indicating that the access volume ratio of this camera during the morning and evening rush hours reaches 80%; at the same time, the "011" code is extracted from the 5th row, representing that its monitoring coverage range is a radius of 150 meters. Subsequently, the spatio-temporal feature data is converted into an 8-bit binary format and subjected to bit interleaving processing, and finally a 16-bit spatio-temporal feature identifier 0x6A57 is generated. In the video data encryption link, the system reserves a special area at the head of the H.264 video stream to embed the spatio-temporal feature identifier, and at the same time backs up the original header information to the extended SEI area and attaches a CRC-8 check code to ensure data integrity.

[0154] This solution brings three significant benefits: in terms of coding efficiency, the storage space requirement for spatio-temporal features is reduced by using the Huffman compression algorithm; in terms of security, the spatio-temporal feature identifier generated by bit interleaving has extremely high anti-tampering characteristics, improving its security strength; in terms of actual performance, the innovative head embedding design only causes an extremely small increase in latency, almost not affecting the smoothness of real-time monitoring. The actual deployment data shows that this solution can improve the accuracy of identifying abnormal access behaviors, including security risks such as unauthorized video viewing during non-authorization periods, and at the same time perfectly supports the dynamic access control strategy based on geographical location, providing reliable data security protection for key areas such as smart city management and border security.

[0155] Figure 2 The following is a schematic structural diagram of a dynamic encryption system for geographical information data based on spatio-temporal features provided by the embodiments of the present application, as Figure 2 shown, this system includes: A determination module 21, configured to determine a dynamic hot spot area according to the timestamp data, geographical coordinate data, and access frequency data of the user access request in the multi-institution sharing scenario; The first generation module 22 is configured to generate a multi-dimensional access weight matrix according to the time distribution density and the spatial coverage radius of the dynamic hot spot area, where the multi-dimensional access weight matrix includes a time dimension decay factor and a spatial dimension distance factor; The second generation module 23 is configured to analyze the access heat values of the spatial grids in the multi-dimensional access weight matrix, determine the corresponding block granularity level according to the numerical range of the access heat values, and perform a spatial segmentation operation on the target geographic information data according to the block granularity level to generate a plurality of data blocks matching the spatial grids; The third generation module 24 is configured to generate a dynamic key seed based on the time dimension decay factor and the spatial dimension distance factor; The encryption module 25 is configured to perform a discrete logarithm operation on the dynamic key seed to generate a temporary encryption key, perform a homomorphic encryption operation on the plurality of data blocks using the temporary encryption key, and embed a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix in the encrypted data blocks; The processing module 26 is configured to, when receiving a cross-institutional data aggregation request, extract the permission level information in the spatio-temporal feature identifier from the encrypted data blocks, perform a ciphertext superposition operation on the plurality of encrypted data blocks based on the homomorphic encryption characteristic, and simultaneously verify the matching degree between the user identity identifier of the requestor and the permission level information to complete the secure aggregation operation.

[0156] Figure 2 The described dynamic encryption system for geographic information data based on spatio-temporal features can execute Figure 1 The described dynamic encryption method for geographic information data based on spatio-temporal features in the illustrated embodiment, and its implementation principle and technical effects will not be elaborated. For the dynamic encryption system for geographic information data based on spatio-temporal features in the above embodiment, the specific manners in which each module and unit perform operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0157] In a possible design, Figure 2 The dynamic encryption system for geographic information data based on spatio-temporal features in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown, and this computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, where the one or more computer instructions are called and executed by the processing component 32.

[0158] The processing component 32 is used for the Figure 1 described dynamic encryption method for geographic information data based on spatio-temporal features in the above embodiment.

[0159] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-mentioned method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above-mentioned method.

[0160] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0161] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0162] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above-mentioned peripheral interface module may be an output device, an input device, etc.

[0163] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0164] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above-mentioned processing component, storage component, etc. may be basic server resources leased or purchased from the cloud computing platform.

[0165] The embodiment of the present application also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above-mentioned Figure 1 A method for dynamically encrypting geographic information data based on spatio-temporal features shown in the above embodiment.

[0166] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0167] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative effort.

[0168] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0169] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A dynamic encryption method for geographic information data based on spatio-temporal features, characterized in that, Including: Determine a dynamic hot spot area based on the obtained timestamp data, geographical coordinate data, and access frequency data of user access requests in a multi-institution sharing scenario; Generate a multi-dimensional access weight matrix according to the time distribution density and spatial coverage radius of the dynamic hot spot area, where the multi-dimensional access weight matrix includes a time dimension decay factor and a spatial dimension distance factor; Analyze the access heat values of spatial grids in the multi-dimensional access weight matrix, determine the corresponding block granularity level according to the numerical range of the access heat values, and perform a spatial segmentation operation on the target geographical information data according to the block granularity level to generate multiple data blocks matching the spatial grids; Generate a dynamic key seed based on the time dimension decay factor and the spatial dimension distance factor; Perform a discrete logarithm operation on the dynamic key seed to generate a temporary encryption key, use the temporary encryption key to perform a homomorphic encryption operation on the multiple data blocks, and embed a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix in the encrypted data blocks; When receiving a cross-institution data aggregation request, extract the permission level information in the spatio-temporal feature identifier from the encrypted data blocks, perform a ciphertext superposition operation on the multiple encrypted data blocks based on the homomorphic encryption property, and simultaneously verify the matching degree between the user identity identifier of the requestor and the permission level information to complete the secure aggregation operation.

2. The method according to claim , wherein, The analyzing the access heat values of spatial grids in the multi-dimensional access weight matrix, determining the corresponding block granularity level according to the numerical range of the access heat values, and performing a spatial segmentation operation on the target geographical information data according to the block granularity level to generate multiple data blocks matching the spatial grids includes: Mark the initial access heat value within the coverage range of each spatial grid in the multi-dimensional access weight matrix, where the initial access heat value is the product of the time distribution density and the spatial coverage radius; Construct a heat diffusion window based on the adjacent relationship between spatial grids, and perform horizontal propagation and accumulation of the initial access heat value within the heat diffusion window to obtain the access heat value; Input the access heat value into a preset discretized step function to divide the corresponding block granularity level according to each step interval through the discretized step function; Determine the maximum number of blocks allowed within the spatial grid according to the block granularity level, and perform radial cutting at equal angular intervals along the longitude and latitude directions with the center point of the spatial grid as the reference to form multiple initial data blocks; When the cutting lines of the multiple initial data blocks overlap with the adjacent spatial grids, re-divide the boundaries of each initial data block with the center line of the overlapping area as the boundary to generate multiple data blocks, and each data block belongs to a single spatial grid.

3. The method according to claim 1, characterized in that, Generating a dynamic key seed based on the time dimension decay factor and the spatial dimension distance factor includes: Perform a timestamp difference operation based on the time dimension decay factor to generate a time decay coefficient; Perform a geographical coordinate difference operation based on the spatial dimension distance factor to generate a spatial distance coefficient; Input the time decay coefficient and the spatial distance coefficient into a non-linear transformation function to generate a dynamic key seed.

4. The method according to claim 1, wherein Perform a discrete logarithm operation on the dynamic key seed to generate a temporary encryption key, perform a homomorphic encryption operation on the multiple data blocks using the temporary encryption key, and embed a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix in the encrypted data block, including: Perform a cyclic shift operation on the initial value of the dynamic key seed, perform an exclusive OR operation on the element corresponding to each initial value and the adjacent element after shifting to generate an extended key sequence, and select elements that meet the preset parity condition based on the parity distribution characteristics of the elements in the extended key sequence to generate a temporary encryption key; Divide the temporary encryption key into multiple sub-key segments, each sub-key segment corresponding to the encryption position of a data block. Calculate the starting offset of the sub-key segment in the temporary encryption key sequence according to the spatial grid coordinates of the data block, and intercept a sub-key segment of the corresponding length according to the offset as the encryption parameter of the data block to perform a homomorphic encryption operation on the multiple data blocks to generate encrypted data blocks; Bit-cross weave the time distribution density feature code and the spatial coverage radius feature code of the spatial grid from the multi-dimensional access weight matrix to generate a spatio-temporal feature identifier, and embed the spatio-temporal feature identifier into the corresponding encrypted data block.

5. The method according to claim 1, wherein When receiving a cross-institutional data aggregation request, extract the permission level information in the spatio-temporal feature identifier, perform a ciphertext superposition operation on multiple encrypted data blocks based on the homomorphic encryption feature, and simultaneously verify the matching degree between the user identity identifier of the requester and the permission level information to complete the secure aggregation operation, including: When receiving a cross-institutional data aggregation request, extract the spatio-temporal feature identifier from the encrypted data block, and extract the permission level information composed of a digital sequence from a fixed position of the spatio-temporal feature identifier. Compare the permission level information item by item with the pre-stored institutional permission mapping table, and screen out multiple target encrypted data blocks that are allowed to participate in the superposition operation from multiple encrypted data blocks; Divide the user identity identifier of the requester carried in the cross-institutional data aggregation request into a head verification code and a tail permission identifier. The head verification code is converted into a binary mask through a preset permutation rule and performs a bit coverage match with the verification area of the spatio-temporal feature identifier; the tail permission identifier inversely deduces the theoretical identity feature identifier through the permission level information. When the continuous matching times between the theoretical identity feature identifier and the verification bit of the user identity identifier reach a preset threshold, activate the superposition operation channel; Separate the multiple target encrypted data blocks into multiple ciphertext segments separately, and input the ciphertext segments with the same position index in the multiple encrypted data blocks into the superposition operation channel. Generate a dynamic weight factor based on the permission level information corresponding to the multiple encrypted data blocks, and perform weighted multiplication accumulation on the ciphertext segments in the same channel to generate the ciphertext segment at the corresponding position in the aggregated ciphertext; During the superposition process, count in real time the number of ciphertext segments that have been superposed in the aggregated ciphertext for each encrypted data block. If the superposition number of any target encrypted data block exceeds the maximum superposition segment number threshold corresponding to the permission level information of the target encrypted data block, close the superposition operation channel corresponding to the target encrypted data block.

6. The method according to claim 3, wherein Performing timestamp difference operation based on the time dimension decay factor to generate a time decay coefficient, including: Recording the first marked timestamp value at the starting moment of the event and the second marked timestamp value at the current moment of the event using the same time measurement rule; Performing a timestamp difference operation on the first marked timestamp value and the second marked timestamp value to obtain an initial time difference, and dividing the initial time difference by the corresponding value of the preset basic time unit to obtain a standard time interval; Extracting the predefined time dimension decay factor, using the time dimension decay factor as the base number, and the standard time interval as the exponent, performing the exponentiation operation of the base number to generate a time decay coefficient.

7. The method according to claim 4, characterized in that Cross-weaving the time distribution density feature code and the spatial coverage radius feature code of the spatial grid from the multi-dimensional access weight matrix to generate a spatio-temporal feature identifier, and embedding the spatio-temporal feature identifier into the corresponding encrypted data block, including: Extracting the time distribution density feature code of the spatial grid from the first predefined column of the multi-dimensional access weight matrix, and extracting the spatial coverage radius feature code of the spatial grid from the second predefined row; Converting the time distribution density feature code and the spatial coverage radius feature code into binary sequences with a fixed length respectively, and sequentially cross-stitching each bit of the two binary sequences according to the alternating arrangement rule to generate a spatio-temporal feature identifier; Reserving an embedding area with a fixed length at the head of the encrypted data block, and filling the spatio-temporal feature identifier into the embedding area bit by bit to cover the values of the original data bits.

8. A dynamic encryption system for geographic information data based on spatio-temporal features, characterized in that, Including: A determination module for determining a dynamic hot spot area according to the timestamp data, geographical coordinate data, and access frequency data of the user access request in the multi-institution sharing scenario; A first generation module for generating a multi-dimensional access weight matrix according to the time distribution density and spatial coverage radius of the dynamic hot spot area, where the multi-dimensional access weight matrix includes a time dimension decay factor and a spatial dimension distance factor; A second generation module for parsing the access heat value of the spatial grid in the multi-dimensional access weight matrix, determining the corresponding block granularity level according to the numerical range of the access heat value, and performing a spatial segmentation operation on the target geographical information data according to the block granularity level to generate multiple data blocks matching the spatial grid; A third generation module for generating a dynamic key seed based on the time dimension decay factor and the spatial dimension distance factor; An encryption module for performing a discrete logarithm operation on the dynamic key seed to generate a temporary encryption key, using the temporary encryption key to perform a homomorphic encryption operation on the multiple data blocks, and embedding a spatio-temporal feature identifier associated with the multi-dimensional access weight matrix into the encrypted data block; A processing module for, when receiving a cross-institution data aggregation request, extracting the permission level information in the spatio-temporal feature identifier from the encrypted data block, performing a ciphertext superposition operation on the multiple encrypted data blocks based on the homomorphic encryption characteristic, and simultaneously verifying the matching degree between the user identity identifier of the requestor and the permission level information to complete the secure aggregation operation.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for dynamically encrypting geographic information data based on spatio-temporal features as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, it implements a method for dynamically encrypting geographic information data based on spatio-temporal features as described in any one of claims 1 to 7.

Images